3.5 Real-time results

QPEV o ið Þ ;t � �<sup>2</sup>

Research Trends and Challenges in Smart Grids

≤

SOCF

where the final SOC reached in Stage (I), that is, SOCR

V2 ð Þ i;t Xð Þ<sup>i</sup>

!<sup>2</sup>

QPEV o ið Þ;<sup>t</sup> þ

achieved SOC, that is, SOCF

chð Þ<sup>i</sup> ð Þ;<sup>t</sup> :

3.3.2 Problem formulation of stage (II)

except for (47), which is replaced by

3.3.3 Problem formulation of stage (III)

3.4 Coordination between V2GQ and COC

Stage (II), that is, P<sup>R</sup>

66

SOCF

Thus, the objective function of Stage (II) is

where SPEV

owners SOCD

≤ SPEV o ið Þ ;t � �<sup>2</sup>

Vmax c ið Þ <sup>V</sup>ð Þ <sup>i</sup>;<sup>t</sup> Xð Þ<sup>i</sup> � �<sup>2</sup>

chð Þ<sup>i</sup> ð Þ;<sup>t</sup> <sup>≤</sup> SOCD

chð Þ<sup>i</sup> ð Þ;<sup>t</sup> <sup>¼</sup> SOCR

max<sup>γ</sup> <sup>∑</sup> i∈IDG

restore a feasible solution for the COC and relax the tap operation. Thus,

min<sup>γ</sup> <sup>∑</sup> i∈I<sup>b</sup>

o ið Þ ;<sup>t</sup> , should be maintained.

Po ið Þ ;<sup>t</sup> <sup>¼</sup> PR

� PPEV o ið Þ;t � �<sup>2</sup>

> � <sup>P</sup>PEV o ið Þ ;t � �<sup>2</sup>

o ið Þ ;<sup>t</sup> is the rated power of the PEV converter. In addition, the final

In Stage (II), the objective is to minimize the DG active power curtailment,

ensure maximum customer satisfaction, which is the highest priority of the V2GQ technology. Therefore, this stage is subject to all of the constraints in Stage (I)

Stage (III) aims at minimizing the voltage deviation using the DGs and PEVs to

1 � Vð Þ <sup>i</sup>;<sup>t</sup> � �<sup>2</sup>

Besides all the constraints in Stage (II), this problem is subject to the constraint defined in (50), in which the maximum injected powers from the DGs reached in

The charging decisions and active/reactive dispatch signals produced in Stage (III) are sent to all PEV parking lots and DGs, as shown in Figure 9. To ensure that the PEV and DG converters settle at the desired active and reactive power references, a time delay Δtconv is introduced. The converter settling time may vary from 50 to 100 ms, depending on the primary controllers of the DC/AC converter [7]. For slow automatic interactions, such as voltage regulation, the maximum communication time delay is 100 ms as per the IEC 61850 [20]. Thus, Δtconv is assumed to be

chð Þ<sup>i</sup> ð Þ;<sup>t</sup> , should be limited by the SOC desired by the PEV

chð Þ<sup>i</sup> ð Þ;<sup>t</sup> , <sup>∀</sup><sup>i</sup> <sup>∈</sup>IPEV, chð Þ<sup>i</sup> , t (47)

chð Þ<sup>i</sup> ð Þ;<sup>t</sup> , <sup>∀</sup><sup>i</sup> <sup>∈</sup><sup>I</sup> PEV, chð Þ<sup>i</sup> , t (48)

Po ið Þ ;<sup>t</sup> , ∀t (49)

, ∀i, t (50)

o ið Þ ;<sup>t</sup> , <sup>∀</sup>i∈<sup>I</sup> DG, t (51)

, ∀i∈I PEV, t (45)

, ∀i ∈I PEV, t (46)

chð Þ<sup>i</sup> ð Þ;<sup>t</sup> , must be attained to

In this section, various case studies are presented to validate the robustness and effectiveness of the optimal coordinated voltage regulation algorithm. The 38-bus 12.66-kV distribution system is used as a test system, as shown in Figure 10. The system data can be found in [21]. The system is modified to accommodate two PEV parking lots and four PV-based DGs, with power ratings as given in Figure 10. The power demands of the two parking lots are extracted from data provided by the TPA for a weekday in 2013. Both parking lots are commercial, where P1 represents a lot in the vicinity of a train station and P2 is a lot located near downtown Toronto. The total number of PEVs during a day is displayed in Figure 11. Due to confidentiality, the addresses of the real parking lots are not mentioned. The central control unit receives the desired SOCs and sends the charging decisions to all vehicles in the parking lots. An OPAL real-time simulator (RTS) is used to model the visual test network using the SimPowerSystems blockset, which is available in Simulink/ Matlab, and an ARTEMiS plug-in [22]. The network, PEV, and DG models are

Figure 10. Test network with an HiL realization.

impacts of the OLTC. Figure 12(a) and (b) illustrate the response of the conventional controller for the OLTC over 24 hours. Although the OLTC does not suffer from an excessive tap operation (13 taps/day), undervoltage and overvoltage prob-

As expected, the overvoltage happens during the peak power generation from DGs, while the undervoltage coincides with the peak demand of PEV. The COC is enabled to mitigate these voltage violations. Figure 12(c) and (d) demonstrate that the COC can limit the voltage violations without the PEV/DG voltage support, but it suffers from a hunting problem. This problem happens when the overvoltage and undervoltage occur concurrently. In this situation, the COC should be deactivated.

To address the hunting problem, presented in the previous case, the V2GQ is coordinated with the COC. Two case studies dealing with PEV/DG voltage support

1. DG active power curtailment, without PEV and DG reactive power supports,

Figure 13 demonstrates the performance of the coordination algorithm when the voltage support is merely achieved via the DG active power curtailment (i.e., the

o ið Þ ;<sup>t</sup> <sup>¼</sup> <sup>0</sup>, <sup>∀</sup><sup>i</sup> <sup>∈</sup>IPEV and Qo ið Þ;<sup>t</sup> <sup>¼</sup> <sup>0</sup>, <sup>∀</sup><sup>i</sup> <sup>∈</sup><sup>I</sup> DG

o ið Þ ;<sup>t</sup> ; Po ið Þ ;<sup>t</sup> ; Qo ið Þ ;<sup>t</sup>

h i

3.5.2 OLTC control with PEV/DG voltage support

h i, where <sup>Q</sup>PEV

Response of coordination algorithm, assuming DG active power curtailment.

2. PEV and DG reactive power dispatch, that is, <sup>γ</sup> <sup>¼</sup> <sup>X</sup> chð Þ<sup>i</sup> ð Þ;<sup>t</sup> ; <sup>Q</sup>PEV

are carried out, as follows:

Voltage Regulation in Smart Grids

DOI: http://dx.doi.org/10.5772/intechopen.85108

that is, <sup>γ</sup> <sup>¼</sup> <sup>X</sup> chð Þ<sup>i</sup> ð Þ;<sup>t</sup> ; Po ið Þ ;<sup>t</sup>

lems occur.

Figure 13.

69

Figure 11. Number of vehicles in the parking lots.

distributed between the RTS cores for performing parallel computations. The RTS is used to perform a hardware-in-the-loop (HiL) realization, where a central control unit, emulated by a host computer running GAMs, exchanges real-time data with the test network modeled in the RTS. The sampling time used to realize the HiL application is 100 μs.

### 3.5.1 OLTC control without PEV/DG voltage support

This section compares the responses of the conventional and COC controllers for the OLTC. The voltage support from PEVs and DGs is disabled to study their

Figure 12. OLTC response: (a & b) conventional control and (c & d) COC.
